Changes between Version 29 and Version 30 of ClusterDMtg070209

07/28/09 17:38:12 (12 years ago)



  • ClusterDMtg070209

    v29 v30  
    151151        1:30            Measurement project  [  wiki] [  slides] (Keren Bergman, Franz Fidler)
     153Discussion on monitoring
     155- Jeff: hyperic may be a good tool that uses ganglia to monitor VM
     156  status, can be extended to other things than VM; have a student
     157  working on this; looking for input as to whether this is a good
     158  direction.
     160  Ilia: have looked at PCP for a monitoring system.  some concern that
     161  Hyperic may have an uncertain future as it was acquired last month.
     162  have indicated to GMOC that there is a strong preference to
     163  developing a SOAP interface
     165  Heidi: need to think about what data you want to expose and share
     167  Anish: Kansei has an interface for getting experiment and substrate
     168  status. 
    153171        2:00            Discussion of Cluster D integration thru end of Spiral 1
    154172                        Current summary view, combining all projects  (see attached word document [  ] and power point file [  ])[[BR]]
    164182                        How close are we to consensus?
     184Discussion of Spiral 2 Capabilities  (see Harry's notes)
     186- Chip: central goal of Spiral 2 is live experimentation; won't be
     187  easy; everything else is secondary
     189- Experiments: Want to see experiments of end-to-end systems, e.g.,
     190  matching networks with servers, sensors with clouds, content with
     191  buses, radars & buses
     193- Identity management: DOME plans to plug into some foreign
     194  authentication manager.  GPO plans to try to go to Shib & InCommon,
     195  since it looks like it might take hold broadly.  Rough mental model
     196  is that in the early days, almost anybody can use this.  Need ways
     197  to say 'these people are out' but probably don't need elaborate
     198  policies early on.  GPO is interested in knowing whether folks
     199  believe moving to Shibboleth is a bad idea.  Shib has been most
     200  successful for managing user access to a portal, seems good for GENI
     201  as far as that goes and that's what ORCA uses it for.  What that
     202  user is allowed to do is another issue and not currently handled by
     203  Shib attributes.  Shib is only starting to handle delegated
     204  authority, still early days.
     206- Improved integration. 
     208  Need to improve distribution of keys to containers.  Currently
     209  manual, ugly, error prone, requires documentation but will work
     210  until container reaches 10's.  Currently no privacy.  Could be added
     211  by using HTTPS on transport.
     213- Measurements
     215  DOME collects location information on buses but needs a solution on
     216  providing it to GENI users.  Has also been thinking about
     217  longitudinal studies, Brian will send Chip a copy of a paper.
     219  ERTM will prototype hardware, interface it with BEN, integrate with
     220  ORCA.
     222- Interoperability.  Want to avoid CF balkanization.
    166224        4:00            Adjourn
    169 As you can see, we need to stick closely to this agenda if we are to finish with the information we need.  And, PLEASE bring no more than 6 slides for each project. 
    172 Please send me your comments or questions.  I am looking forward to our meeting on Thursday!
    175 Thanks,
    177 Harry